Systems for sorting seeds are disclosed, as well as batches of seeds that have been sorted using the systems.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
2. The system according to claim 1, wherein the similar extractable at least one visual feature is selected from the group consisting of: a hand-crafted feature, at least one size dimension of the at least one seed, color of the at least one seed, shape of the at least one seed, and texture of the at least one seed.
3. The system according to claim 1, wherein the at least one image includes a plurality of seeds that differ from one another within a tolerance range by a single feature that cannot be extracted by the at least one visual feature, and further comprising computing clusters according to a respective binary classification category computed for each seed, wherein the respective binary classification category is indicative of the respective seed including the single feature or not including the single feature, and wherein the instructions include instructions for sorting the seeds according to the computed clusters.
This invention relates to a system for analyzing images to identify and classify specific features that cannot be extracted using conventional visual feature extraction techniques. The system addresses the challenge of detecting subtle differences between similar objects or elements (referred to as "seeds") in an image, where traditional feature extraction methods fail to distinguish them. The system processes at least one image containing multiple seeds that differ from one another by a single, hard-to-extract feature. These seeds are grouped into clusters based on a binary classification category, which determines whether each seed includes or lacks the single feature. The classification is performed by analyzing the seeds and assigning them to clusters accordingly. The system then sorts the seeds based on these computed clusters, enabling precise differentiation and organization of the seeds based on the presence or absence of the single feature. This approach enhances the ability to detect and categorize subtle differences in images, improving applications such as quality control, object recognition, and pattern analysis where traditional visual features are insufficient. The system leverages binary classification to overcome limitations in feature extraction, providing a more refined and accurate method for distinguishing similar elements in an image.
4. The system according to claim 3, wherein the binary classification category indicative of the single feature is selected from the group consisting of: self pollinated or hybrid pollinated, stress-resistant or non-stress resistant, genetically modified or non-genetically modified, isogenic seeds that differ by the single feature, and seeds of a shared mother plant with different paternal pollen.
This invention relates to a system for classifying seeds based on specific binary features. The system is designed to address challenges in seed selection and breeding by providing a method to categorize seeds into distinct groups based on a single genetic or phenotypic trait. The system analyzes seeds to determine whether they fall into one of several predefined binary categories, including self-pollinated versus hybrid-pollinated, stress-resistant versus non-stress-resistant, genetically modified versus non-genetically modified, isogenic seeds differing by a single feature, or seeds from a shared mother plant with different paternal pollen. The classification process involves evaluating the seeds to identify the presence or absence of the target feature, enabling precise categorization. This system supports agricultural research, breeding programs, and seed production by streamlining the identification of seeds with desired traits, improving efficiency in plant breeding and genetic studies. The invention enhances the ability to track and select seeds based on specific characteristics, facilitating advancements in crop development and genetic diversity management.
5. The system according to claim 1, wherein the at least one classification category comprises at least one member selected from the group consisting of: (i) a non-visual category that cannot be manually determined based on visual inspection of the at least one seed, (ii) a seed variant, (iii) not directly correlated to DNA markers, (iv) a yield-related trait, (v) a monogenetic trait, (vi) a pleiotropic trait, (vii) a polygenetic trait, a (viii) plant quality related trait, (ix) a genotype, and (ix) at least one phenotypical property predicted to develop in the at least one seed at a future time interval relative to a time interval when the at least one image is captured.
6. The system according to claim 1, wherein the indication of the at least one classification category associated with respective plurality of training images of the training dataset comprises a seed variant determined according to a parent plant.
This invention relates to a plant classification system that uses image-based training data to categorize plants. The system addresses the challenge of accurately classifying plants by leveraging a structured training dataset where each plant image is associated with a classification category. A key feature is the use of a "seed variant" to determine the classification category for a group of training images. The seed variant is derived from a parent plant, ensuring that the classification is based on genetic or phenotypic consistency. The system processes multiple training images, each linked to a specific classification category, and uses the seed variant to refine or validate the categorization. This approach improves the accuracy of plant classification by incorporating genetic relationships, which is particularly useful in agriculture, botany, and automated plant identification systems. The system may also include preprocessing steps to enhance image quality or extract relevant features before classification. The overall goal is to provide a robust method for categorizing plants based on visual data while accounting for genetic lineage.
7. The system according to claim 1, wherein the indication of the at least one classification category associated with respective plurality of training images of the training dataset is based on a DNA test destructive to the seed from which it was obtained.
8. The system according to claim 1, wherein the at least one neural network computes an embedding for the at least one image, and wherein the at least one classification category is determined according to an annotation of an identified at least one similar embedded image from the training dataset storing embeddings of training images, the at least one similar embedded image identified according to a requirement of a similarity distance between the embedding of the at least one image and embedding of the training images.
9. The system according to claim 8, wherein the embedding is computed by an internal layer of the trained at least one neural network selected as an embedding layer.
10. The system according to claim 8, wherein the embedding is stored as a vector of a predefined length, wherein the similarity distance is computed as a distance between a vector storing the embedding of the at least one image and a plurality of vectors each storing embedding of respective training images.
11. The system according to claim 8, wherein the similarity distance is computed between the embedding of the at least one image and a cluster of embeddings of a plurality of training images each associated with a same at least one classification category.
12. A container comprising a plurality of seeds, wherein said plurality of seeds are sorted according to the system of claim 1.
13. The container of claim 12, wherein said plurality of seeds are identical with respect to a trait, a microbiome or a genome.
14. The container of claim 12, wherein at least one member is selected from the group consisting of: (i) wherein said plurality of seeds comprises more than 1000 seeds, and (ii) wherein said plurality of seeds weights more than 100 grams.
15. The container of claim 12, wherein said trait is selected from the group consisting of increased nitrogen use efficiency, increased abiotic stress tolerance, increased biotic stress tolerance, increased biomass, increased growth rate, increased vigor, increased yield and increased fiber yield or quality, and increased oil.
16. A method of growing a crop comprising seeding the seeds of the container of claim 12, thereby growing the crop.
17. The system of claim 1, wherein a statistical classifier trained for extraction of the at least one visual feature classifies the plurality of seeds which have similar extractable at least one visual feature into a same classification category for which visual features are explicitly defined.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
December 3, 2018
November 22, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.